Disturbance Estimation based Robust Consensus of Nonlinear Multi-agent Systems

In this work, a new design based on disturbance estimation for the robust consensus of multi-agent systems is proposed where all agents are modelled as non-linear dynamic systems. Here a fixed topology is considered and it is assumed that all agents share relative information of their states with each other. The average consensus algorithm is augmented with Uncertainty and Disturbance Estimator (UDE) based disturbance estimation to achieve robustness. In doing so, state dependent nonlinearities of the system are considered as a part of the uncertainties to be estimated. The proposed robust consensus has two aspects, the agreement of the states of all agents and the estimation as well as cancellation of external disturbances acting on each agent. Here estimation of disturbances acting on the system is obtained using novel filter design of UDE. Closed loop stability of the overall system is established and numerical simulation results of proposed approach are presented to demonstrate its efficacy against different disturbances.